fofr / lcm-animation

Fast animation using a latent consistency model

  • Public
  • 30.7K runs
  • L40S
  • GitHub
  • License
Iterate in playground

Input

string
Shift + Return to add a new line

Prompt to start with, if not using an image

Default: "Self-portrait oil painting, a beautiful cyborg with golden hair, 8k"

string
Shift + Return to add a new line

Prompt to animate towards

Default: "Self-portrait watercolour, a beautiful cyborg with purple hair, 8k"

file

Starting image if not using a prompt

integer

Width of output. Lower if out of memory

Default: 512

integer

Height of output. Lower if out of memory

Default: 512

integer

Number of times to repeat the img2img pipeline

Default: 12

number
(minimum: 0, maximum: 1)

Prompt strength when using img2img. 1.0 corresponds to full destruction of information in image

Default: 0.2

integer
(minimum: 1, maximum: 50)

Number of denoising steps. Recommend 1 to 8 steps.

Default: 8

boolean

Use canny edge detection to guide animation

Default: true

number
(minimum: 0.1, maximum: 4)

Controlnet conditioning scale

Default: 2

number
(minimum: 0, maximum: 1)

Controlnet start

Default: 0

number
(minimum: 0, maximum: 1)

Controlnet end

Default: 1

number
(minimum: 1, maximum: 255)

Canny low threshold

Default: 100

number
(minimum: 1, maximum: 255)

Canny high threshold

Default: 200

integer
(minimum: 0, maximum: 4)

Zoom increment percentage for each frame

Default: 0

number
(minimum: 1, maximum: 20)

Scale for classifier-free guidance

Default: 8

integer

Random seed. Leave blank to randomize the seed

boolean

Return a tar file with all the frames alongside the video

Default: false

Output

Generated in

This output was created using a different version of the model, fofr/lcm-animation:9f10cbd0.

Run time and cost

This model costs approximately $0.052 to run on Replicate, or 19 runs per $1, but this varies depending on your inputs. It is also open source and you can run it on your own computer with Docker.

This model runs on Nvidia L40S GPU hardware. Predictions typically complete within 54 seconds. The predict time for this model varies significantly based on the inputs.

Readme

This model doesn't have a readme.